What is mcp-token-tracker CLAUDE.md?

A command-line program that scans Model Context Protocol settings and estimates how many language-model tokens loaded tool descriptions use. It can also report estimated monthly cost and recorded server usage when call logging is available.

Use it to inspect MCP configurations from Claude Desktop, Claude Code, Cursor, or Windsurf and review token counts, estimated costs, and usage by server.

It shows which connected tool servers consume context and may cost money. This helps you decide which servers should remain loaded.

How forbiddenlink/mcp-token-tracker configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to forbiddenlink/mcp-token-tracker. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/forbiddenlink/mcp-token-tracker/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/forbiddenlink/mcp-token-tracker

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/forbiddenlink/mcp-token-tracker/claude-md/github.svg)](https://agentmods.dev/instructions/forbiddenlink/mcp-token-tracker/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/forbiddenlink/mcp-token-tracker/claude-md"><img src="https://agentmods.dev/badge/instructions/forbiddenlink/mcp-token-tracker/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/forbiddenlink/mcp-token-tracker/claude-md"><img src="https://agentmods.dev/badge/instructions/forbiddenlink/mcp-token-tracker/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,204 This file is loaded in full into every session.

When invoked 1,204 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.01204	$0.01204
Opus 5.5	$0.00482	$0.00482
Sonnet 5.5	$0.00241	$0.00241
Haiku 4.5	$0.00120	$0.00120

Measured 1mo ago against content hash bf61ce15de62, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mcp-token-tracker CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

mcp-token-tracker

CLI that scans MCP (Model Context Protocol) configs on disk and estimates the token cost of loaded tool definitions, to help decide which MCP servers to keep loaded. Repo: https://github.com/forbiddenlink/mcp-token-tracker

What it does

Scans known MCP config locations (Claude Desktop, Claude Code, Cursor, Windsurf), counts tool definition tokens with tiktoken, estimates monthly cost, and reports per-server usage (call counts, last-used) when the Claude Code hook has been logging calls.

Stack

TypeScript, strict mode, ESM ("type": "module"), target ES2022
Node 20 (.nvmrc)
pnpm 10.34.5 (pinned via packageManager)
Commander.js (CLI), chalk (color output), boxen (config display), conf (persisted user config)
tiktoken for token counting, @modelcontextprotocol/sdk for live server connections (--live)
pino for structured logging
Langfuse (optional, only active if LANGFUSE_PUBLIC_KEY/LANGFUSE_SECRET_KEY are set)

Commands

pnpm install
pnpm dev              # tsx src/cli.ts scan
pnpm build            # tsc -> dist/
pnpm start            # node dist/cli.js
pnpm check            # build + test
pnpm test             # vitest run
pnpm lint-baseline    # biome check .
pnpm audit            # pnpm audit --audit-level high

CLI usage: pnpm dev (fast estimate), pnpm start scan --json (machine-readable), npx tsx src/cli.ts scan --live (connects to servers for real tool definitions).

Layout

src/cli.ts - Commander entry point, scan command
src/scanner.ts - finds and parses MCP config files
src/analyzer.ts - token counting and cost estimation
src/connector.ts - live JSON-RPC connection to MCP servers (--live mode)
src/usage.ts - reads/writes ~/.mcp-token-tracker/usage.json (call history, capped at 10,000 entries / 90 days)
src/lib/config.ts - persisted user config via conf (budgets, per-model cost tables)
src/lib/logger.ts - pino logger + child loggers per module
src/lib/langfuse.ts - optional Langfuse client
src/trigger/ - empty except .gitkeep; see Gotchas
*.test.ts files are colocated with the source they test (vitest)

Read the full file on GitHub · 97 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

1mo ago First seen · 97 lines · 1,204 tokens per session scan A bf61ce15de62

Subscribe to this mod's changes

mcp-token-tracker CLAUDE.md is an instructions file published in the GitHub repository forbiddenlink/mcp-token-tracker (2 stars, last pushed 2d ago), licensed MIT. It adds 1,204 tokens to every session, about $0.0048 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-20.

Analytics and metrics Data and AI TypeScript ai claude cost-tracking developer-tools mcp

mcp-token-tracker CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
mcp-token-tracker CLAUDE.mdforbiddenlink/mcp-token-tracker	Instructions file	2	2d ago	MIT	A	1,204	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

mcp-token-tracker: Instructions file for Claude Code